Understanding the mechanisms by which reactive oxygen species modulate both the expression and the alternative splicing of transcription factors such as KLF6, a tumor suppressor gene mutated in several carcinomas and likely involved in the development of alcoholic liver disease, is of great relevance for potential therapeutic intervention, and this constitutes the major goal of this application. We hypothesize that reactive oxygen species in general and reactive oxygen species derived from cytochrome P450 2E1 in particular may play a critical role in regulating KLF6 splicing and its biological actions. We propose: 1) To explore the significance of CYP2E1-derived reactive oxygen species modulation of KLF6 expression and alternative splicing in CYP2E1-expressing cells and in primary hepatocytes isolated from chronic alcohol fed mice. We plan to analyze the effects of prooxidants and of glutathione depletion and to analyze the contribution of each KLF6 splice isoform to the effects mediated by the prooxidants. 2) To establish the relevance in vivo of the spliced isoforms KLF6_V1 and KLF6_V2 in alcoholic liver disease using the following models: a) wild-type mice infected with lentivirus overexpressing either the full-length KLF6 or any of the spliced variants or the corresponding siRNAs before and after inducing liver injury by ethanol feeding; b) Sod1-/- mice whose compromised antioxidant defense is hoped to increase the responsiveness to reactive oxygen species in term of modulating KLF6 splicing and activity; and c) Ktf6+/- in which the potential heterozygosity may modify the responsiveness to the alcohol effects. 3) To evaluate whether stress-activated kinases modulate the up-regulation and the alternative splicing of KLF6 under reactive oxygen species production in CYP2E1-expresing cells and in hepatocytes from chronic ethanol-fed mice. Mechanistic studies will follow to understand the impact of oxidant stress on KLF6 expression, alternative splicing, and downstream targets such as TNFalpha, TGFbeta, iNOS, catalase, and NFkB. We hope that these models and the approach taken will help us to define how reactive oxygen species can modulate splicing and particularly the splicing of KLF6 and the actions of each isoform to dissect potential therapeutic targets for preventing liver disease.